Copyright - Sproule Academy Corporation 1 Type Curve Analysis for Landmen Introduction 2 Course Objectives Provide an understanding of what type curves are used for and why they are used Understand the key parameters in type curves and how they impact NPV Understand how type curves are created How have type curves evolved with industry completion optimization Using type curves –key issues
51
Embed
Type Curve Analysis for Landmen - Canadian Association …landman.ca/wp/wp-content/uploads/2014/07/Sept-18-2015-Sproule.pdf · Type Curve Analysis for Landmen ... • Decline trend
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Copyright - Sproule Academy Corporation 1
Type Curve Analysis for Landmen
Introduction
2
Course Objectives
� Provide an understanding of what type curves are used for and why they
are used
� Understand the key parameters in type curves and how they impact NPV
� Understand how type curves are created
� How have type curves evolved with industry completion optimization
� Using type curves – key issues
Copyright - Sproule Academy Corporation 2
3
� Binning wells
� Type curve parameter impact on
NPV
� Adjusting for well length
� Evolution of type curves as
completion optimized
� Using type curves
� Introduction
� What are type curves
� Why use type curves
� Key parameters in type curves
� Multi-segment type curve shape
� Work flow to create type curves
Course Modules
4
� All required production data is available in the public domain
� Type curves can be generated from public data – if completion specifics
are required this is usually not available publically
� Vintaging the sample well data sets may be required to address
completion optimization
Data Access
Copyright - Sproule Academy Corporation 3
5
Public Domain Data
� Typically all land, drilling, completion, production and facilities technical
data is available from provincial and national energy regulators
� Data accessed through commercial vendors (IHS, Accumap, Mosaic,
ValNavigator), industry associations (CAPP) and web based services
� Well location, formation tops, drilling depths and completion history
� Log and core data
� Reservoir data
� Well tests
� Well Production
6
� Type curves are estimates based on limited production history and
predictions of future performance
� There is always uncertainty in type curves
� A type curve for a given play is only as reliable as the quality of the data
and the skill of the evaluator
• Garbage In = Garbage Out!
Reliability of Type Curves
Copyright - Sproule Academy Corporation 4
Type Curve Analysis for Landmen
What are type curves?
8
• An idealized well, usually an average result
• Can be accurate
• Widely used for unconventional oil and gas plays
• Not a new technique: used for decades!
• Powerful and useful: all should know how to use!
What’s a Type Curve?
Copyright - Sproule Academy Corporation 5
9
What’s Old is New: Type Curves
10
• Type curve must have a foundation. Examples:
– Performance of analogous pools and wells
• Reserves life index
• Decline profile
– Volumetric calculations
– Normalized well production
– Computer simulations
– Combination of the above
Basis for Type Curves
Copyright - Sproule Academy Corporation 6
11
• Setting well performance to a
common variable, usually start
date
• Can normalize to other
variables. Examples, a fraction
or percentage of:
• Maximum rate
• Maximum recovery
• Gas or oil in place
• On graph to right, normalized
production shows common
trends
Normalizing Well Production
12
Some Things to Consider
• When preparing type curves:
• Geology
• Permeability
• Well type: horizontal
• Drilling and completion techniques
• Number and size of fracs
• Frac design, eg “slickwater”
• And this is only the beginning.
Copyright - Sproule Academy Corporation 7
13
Life Gets Complicated
• When preparing type curves, typically have more, rather than fewer type
curves as time passes.
• Why: mostly geology. Recognize layers, units.
• Completion optimization: ongoing
• Examples:
• Montney gas in NW AB, NE BC:
• 3 layers. 6 areas. 17 curves
• Number and design of fracs optimized
• Pembina Cardium: 3 areas, 63 curves and counting
Type Curve Analysis for Landmen
Why use type curves?
Copyright - Sproule Academy Corporation 8
15
Why use type curves?
• On new plays often the most reliable technique to estimate expected well
performance if sufficient analagous wells exist
• Can be accurate if based on sizeable well sample
• Based on actual well performance vs. volumetrics
• Best representation of well performance and EUR expectations for multi-
stage frac unconventional plays
16
Evolution of Type Well Curves
Increase horizontal well drilling with no production history or analogies. NEEDED A WAY TO FORECAST WELLS
� Original forecast generated from radial flow model or simulator, vertical well results other reservoir analogies
� Over time, had some analogies to use. Still need more time to confirm the “tail”
� Type Well Curves summarize average well performance decline profile to permit effective forecasting and reserve estimation for evaluations
� Better than volumetric’s as it is difficult to define net pay, varying porosity/permeability, Sw and Area
�Better than Energy Balance due to volumetric issues as well as seldom get Rsi, Pi, PVT, defined sweep efficiency, rel. perm ����changing GOR, drive mechanism
Copyright - Sproule Academy Corporation 9
17
Application of Type Curves
• Can use type curves for:
• Waterfloods
• Thermal projects: SAGD, CSS
• Individual Wells
• Multi stage frac unconventional plays
18
Heavy Oil Waterflood Type Curve
0.0
0.2
0.4
0.6
0.8
1.0
0% 5% 10%
Fra
ction o
f Maxim
um
Oil R
ate
Recovery Factor
From type curve profile can estimate:
Primary, secondary, incremental recoveryTime to fill-upDuration of waterflood plateau
Copyright - Sproule Academy Corporation 10
19
Type Curve for SAGD Well Pair
0
4
8
12
16
20
0.0
0.2
0.4
0.6
0.8
1.0
0.06 0.08 0.22 0.37 0.46 0.49
Ste
am
Oil R
atio, Bbl/Bbl
Fra
ction o
f Maxim
um
Oil R
ate
Fraction of Original Oil in Place Recovered
Oil Rate Monthly SOR Cumulative Steam Oil Ratio
Can estimate
production profile,
recovery
20
Heavy Oil Well Type Curve Example
Copyright - Sproule Academy Corporation 11
21
Production Used for Type Curve
22
Example: Kaybob South, Alberta
ERCB ST 98-2011
Variety of type curves modelling performance of
number of treatments
Copyright - Sproule Academy Corporation 12
23
Tying Type Curve to Production
All the Montney horizontal wells in a township in northwest Alberta.The thick line is the average of all the wells’ production.
24
Time-Cum Display For Review
0
1000
2000
3000
4000
0 12 24 36 48 60
Cum
Gas P
roduction MMcf
Months on Production
2 bcf 3 bcf 4 bcf 6 bcf 8 bcf
Copyright - Sproule Academy Corporation 13
25
Estimate Recovery from Type Curve
0
200
400
600
800
1000
1200
1400
0 12 24 36
Cum
ulative G
as P
roduced, MMcf
Months on Production
2 bcf 3 bcf 4 bcf
Montney horizontal well NW Alberta
Month Since First Prod
Cum Gas Prod, MMcf
3 169
6 199
9 409
12 460
18 554
24 638
30 983
36 1020
What’s your answer?
26
Undrilled Locations
• Type curves used extensively to estimate future production forecasts from
individual wells in resource plays
• Used to generate development plan to attain specified facility capacity
levels
• Used to determine ultimate recovery potential (reserves + resources) for a
specific acreage position in a play
• Type well represents the expected average well – represents the best
estimate or 2P or 2C volume and value
Copyright - Sproule Academy Corporation 14
Type Curve Analysis for Landmen
Key parameters in type curves
28
Three Key Parameters
• Initial production rate
• EUR – ultimate recoverable volume
• Curve shape – big impact on NPV10
Copyright - Sproule Academy Corporation 15
29
Initial Production rate
• Strongly influenced by completion technique
• Improvements in the initial rate have been seen in most plays as
completion technique is optimized
• Geology always a big factor
30
EUR
• Ultimate recoverable reserves dependent on
• Geology
• Completion technique
• Well length
• Expect significant variability within a given play
• Typically significant uncertainty in EUR due to limited production history
Copyright - Sproule Academy Corporation 16
31
Curve shape
• Big impact on NPV10
• Often gets the least focus of the three parameters but has huge influence
on value
• Typically three segments in most multi-frac stimulated wells – usually
modeled as two segments for simplicity
• First segment – hypermonic linear flow reflects flow from the fracture
network
• Second segment – exponential or hyperbolic reflects flow from the rock
matrix
• Start of second segment is called time to boundary dominated flow (Tbd)
32
Curve shape - Multi-stage frac
• Initial decline rate
• Time to boundary dominated flow i.e. length of first segment - linear flow
from fracs
• First segment n value (expect 2.0 for linear flow)
• Second segment n value
• Imposition of minimum decline rate for second segment
Copyright - Sproule Academy Corporation 17
33
Curve shape – waterflood
• Reservoir fill up time period
• Peak rate
• Length of plateau period
• Decline trend post plateau – decline rate and n value
34
Curve shape – SAGD Thermal
• Time to reach plateau rate
• Peak rate expectation
• Plateau rate time period
• Post plateau decline rate
• SOR and CSOR expectations – will drive economics
Copyright - Sproule Academy Corporation 18
35
Issues in Type Curve Creation
• recognizing well flow regimes
• Profiles in early time:
• how steep,
• How long
• Long term performance
• updating
0
20
40
60
80
100
0 12 24 36 48 60 72
Oil R
ate
, Bopd
Months on Production
Type Curve for Viking HZ Well
36
Checking Type Curves
• Remember: type curves are an idealization or a model
• Must test model against well performance
• Will have models with a range of performance
• Next slide show verifications of type curves.
• Must continually update type curves as additional performance data is
available
Copyright - Sproule Academy Corporation 19
37
Verifying a Gas Type Curve
Important to check type curves and incorporate production, other information.
Type Curve Analysis for Landmen
Why a multi segment curve shape?
Copyright - Sproule Academy Corporation 20
39
Schematic of fracture
40
Fracture schematic
Copyright - Sproule Academy Corporation 21
41
Flow regimes – vertical vs. hz multi stage frac
42
Multi stage frac nomenclature
Copyright - Sproule Academy Corporation 22
43
Evolution of Type Well Curves - Shape
44
Evolution of Type Well Curves - Shape
Copyright - Sproule Academy Corporation 23
45
Considerations
• Analysis of early time flow
• Look for the “2” but….
• Know that other factors can impact the profile
• Not flowing at constant BH conditions
• Relative permeability effects
• A valid decline curve can have b < 2
• Includes both reservoir and man-made effects
•
46
Second Segment Generation
There s.b. three segments to properly forecast wells but time consuming. Sproule uses two segments (some
competitors use one). The second segment is developed utilizing the following information and experience:
� In house simulation and sensitivities� Decline nature of more mature Hz producing from
similar reservoirs� Studying decline behavior of vertical wells
� Reviewing results of applying realistic declines & “n” to second segment
� Determining how “n” is changing during linear/transition flow
Case 1: The effect of EUR on NPV10 was explored by changing the N on the second segment of the decline curve. The low case is represented by an N2=0, the best case
is N2=0.3 and the high case is N2=0.5.
Case 2: The effect of Qi on NPV10 was explored by using Qi for the low case is 5,000 Mcfd, Qi for the best case is 5,500 Mcfd and the Qi for the high case is 6,000 Mcfd
Case 3: The effect of curve shape on NPV10 was explored by using single segment decline curves for the low and high cases, with N=2 and N=0.5, respectively. The best case is represented by a typical two-segment trend, where the N transitions
from 2 to 0.5 after five years
86
• Overview
Montney Type Curve Sensitivity Case Study
The following case study will highlight the differences in NPV10 that can result when different methods for creating type curves is used, and how too much reliance on
software can skew results. Two cases will be presented:
• Lower Montney AutoThis case represents the resulting type curve if the software scaling function is
used for all individual well declines without any additional user input
• Lower Montney DeclineThis case represents the resulting type curve if all individual well declines are created by the user, and only using the software to calculate the average type
well
All economic inputs required to calculate NPV10 are constant for both cases, and are presented on the next slide.